This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Analyzing *-rtems* ACATS
- From: Joel Sherrill <joel dot sherrill at oarcorp dot com>
- To: GCC List <gcc at gcc dot gnu dot org>
- Date: Fri, 4 Apr 2008 15:07:58 -0500
- Subject: Analyzing *-rtems* ACATS
I posted Ada ACATS results this morning for 4 targets
(i386, powerpc, sparc, and mips) to gcc-testresults.
In general, they were very good -- the powerpc only
had 3 ACATS failures and those were cross target.
Compiler version:
4.4.0 20080403 (experimental) [trunk revision 133868]
The sparc had 6 FPU failures which are simulator deficiencies.
They do not run on sis in gdb but run on tsim.
There are similar FPU issues on the i386 and I have asked
the qemu team if there are known precision issues in qemu.
This accounted for 17 of the 21 failures.
Beyond those, I am left with:
============================================
All targets had the following three failures:
c64005c - "WRONG ITERATIVE TRACE LENGTH."
c64005d - "WRONG ITERATIVE TRACE LENGTH."
c953002 - "WRONG ITERATIVE TRACE LENGTH."
============================================
On the SPARC, there were these which all
all fail with a memory exception after
somehow __gnat_builtin_longjmp
has loaded 0x800 into %fp.
c37213j - unaligned memory access
c37213k - unaligned memory access
c47007a - unaligned memory access
c47009a - partial output, hangs
============================================
Any suggestions on what to investigate for those before
I return to whittling on the C/C++ test results again.
--
Joel Sherrill, Ph.D. Director of Research & Development
joel.sherrill@OARcorp.com On-Line Applications Research
Ask me about RTEMS: a free RTOS Huntsville AL 35805
Support Available (256) 722-9985